*3.5.7 Applying VDC*

VDC analysis aims to measure the size of the contribution or composition of the influence of each variable to other variable. VDC analysis will provide information

Based on **Table 2**, it can be concluded that all variables are stationary at the *first*

a. The FDR variable in the *first difference* level test shows that the ADF t-statistic value is smaller than the *MacKinnon critical value* 5%, which is 11.55782 < 2.890037, which means H0 is rejected and H1 is accepted or the FDR

b. The NPF variable in the *first difference* level test shows that the ADF t-statistic value is smaller than the *MacKinnon critical value* 5%, which is 4.724193 < 2.890623, which means H0 is rejected and H1 is accepted or the NPF

c. The BOPO variable at the first difference level test shows that the ADF t-statistic value is smaller than the MacKinnon critical value 5%, which is 11.03276 < 2.890037, which means H0 is rejected and H1 is accepted or

From the above tests, all variables have met data stationary. The ADF t-statistics are smaller than the McKinnon critical value 5% at the first difference level. Therefore, the next step is to estimate the data by VECM by selecting its lag length criteria.

The lag length is used to determine the effect of the time taken from each variable on the past variable. The selected lag candidates are the length of lag according to the *likelihood ratio* (LR) criterion, *final prediction error* (PPE), *Akaike information criterion* (AIC), *Schwarz information criterion* (SIC), and *Hannan-Quinn criterion* (HQC). The determination of the optimal lag length in this study is based

**Variable t-Statistic The Mackinnon critical value Prob Conclusion 1% 5% 10%** FDR 1.011989 3.495021 2.889753 2.581890 0.7440 Non-stationary NPF 1.55662 3.497029 2.890623 2.582353 0.5009 Non-stationary BOPO 1.786319 3.495021 2.889753 2.58189 0.3854 Non-stationary

**Variable t-Statistic The Mackinnon critical value Prob Conclusion 1% 5% 10%** FDR 11.55782 3.495677 2.890037 2.582041 0.0000 Stationary NPF 4.724193 3.497029 2.890623 2.582353 0.0002 Stationary BOPO 11.03276 3.495677 2.890037 2.582041 0.0000 Stationary

*difference* with a predetermined critical value (α = 5%), as follows:

variable data is stationary.

*Risk Analyses on Islamic Banks in Indonesia DOI: http://dx.doi.org/10.5772/intechopen.92245*

variable data is stationary.

*4.1.2 Lag length criteria*

*Sources: Author's calculation.*

*Sources: Author's calculation.*

*Unit root test-augmented Dickey-Fuller (level).*

*Unit root test-augmented Dickey-Fuller (first difference).*

**Table 1.**

**Table 2.**

**59**

the BOPO variable data are stationary.

**Figure 2.** *Vector error correction model. Source: Gujarati [28].*

about the magnitude and duration of the shock proportion of a variable to the variable itself and to other variables. According to Basuki [26], variance decomposition aims to measure the magnitude of the contribution or composition of the influence of each independent variable on the dependent variable.
